Dynamic Knapsack Model Report

Parameters

Model Parameters

This model is a smaller instance of the full problem:

  • Planning horizon is decreased from 24 weeks to 10 weeks
  • Maximum tracked wait is decreased from 6 weeks to 4 weeks
  • There are 3 surgeries instead of 6 surgeries
  • Number of priorities is set to 1

Simulation Parameters

  • 30 Replications
  • 1250 weeks duration
  • 250 weeks warm up

Surgeries

  1. Surgery 1 - 1. SPINE POSTERIOR DECOMPRESSION/LAMINECTOMY LUMBAR
  2. Surgery 4 - 4. SPINE POST CERV DECOMPRESSION AND FUSION W INSTR
  3. Surgery 6 - 6. SPINE POSTERIOR DISCECTOMY LUMBAR

Arrival Rate

It was set to be 95% of the capacity, however due to transitions, the resource usage should be higher than 95%

Surgery Complexity Arrival_Adjusted Arrival_Original Rationale
Surgery 1 Complexity 1 1.2319885 1.0000 once per week
Surgery 1 Complexity 2 0.6159942 0.5000 once per two weeks
Surgery 4 Complexity 1 0.1437320 0.0833 once per 3 months
Surgery 4 Complexity 2 0.1026657 0.0625 once per 4 months
Surgery 6 Complexity 1 1.2319885 1.0000 once per week
Surgery 6 Complexity 2 0.6159942 0.5000 once per 2 weeks

Resource Usage

Usage of the resources

Surgery Complexity Resource_Type Usage
Surgery 1 Complexity 1 Admissions 0.0
Surgery 1 Complexity 1 OR_Time 3.0
Surgery 1 Complexity 2 Admissions 1.0
Surgery 1 Complexity 2 OR_Time 4.0
Surgery 4 Complexity 1 Admissions 1.0
Surgery 4 Complexity 1 OR_Time 4.0
Surgery 4 Complexity 2 Admissions 1.0
Surgery 4 Complexity 2 OR_Time 5.5
Surgery 6 Complexity 1 Admissions 0.0
Surgery 6 Complexity 1 OR_Time 1.5
Surgery 6 Complexity 2 Admissions 0.0
Surgery 6 Complexity 2 OR_Time 2.5

Resource Capacity

Resouce Capacity_Weekly Unit
Admissions 1.50 Patients Admitted per week
OR_Time 11.25 OR Hours per week

R1R2 Report

This model is the default model that does not penalize bed under utilization

Policy Description

The description will be based on the following graphs (both of them are for M of 0 only):

The graph on the left shows in which days the MDP policy will allow scheduling. If the value is below 0 - the model will allow scheduling, otherwise the model will keep patients in the waitlist. Thus this policy will only schedule into week 1 and week 2

The plot on the right shows the approximate priority of different surgeries. The lower the surgery is on the list the more likely it is to be scheduled first.

Policy Results

R1R2R3 Report

In addition to the parameters above, this model penalizes bed under utilization in the first period

Policy Description

The description will be based on the following graphs:

Policy Results

The policy description is based on the following graphs.